Riding the epidemic curve to glory, Gram-negative edition

Dan has discussed "riding the epidemic curve to glory" before. This is the phenomenom that occurs when a bundle of interventions is started just as an outbreak is entering the “downhill” part of the epidemic curve. Thus, the outbreak would have ended on its own, even without intervention, yet the hospital epidemiologist and their recommended interventions are incorrectly given credit for the success. This effect has only minor importance when looking at a single hospital outbreak. In that situation, it doesn't really matter why the outbreak ended, but it's sure nice that it did.  However, what if the hospital epidemiologist published a paper outlining the successful control of the outbreak? What if others published similar uncontrolled quasi-experimental studies describing how they controlled other outbreaks? Then it might be a problem, no? I think it could be a major problem since we really need to know that interventions work before we recommend them - we can't always rely on the luck of riding the epidemic curve.

This phenomenon is particularly concerning for me when it comes to control of MDR-gram negative infections. A little more background: Back in 2008 we published a paper in ICHE showing that Gram-negative infections were much more common in the summer (vs. winter). In fact, there were 28% more P. aeruginosa, 46% more E. cloacae, 12% more E. coli and 21% more A. baumannii clinical cultures in summer months. We validated these findings in 132 US hospitals and again found that Gram-negative organisms were more frequent in summer months ranging from 12.2% higher rates for E. coli to 51.8% higher for Acinetobacter spp.

Below I've pasted a figure plotting 8-years of monthly aggregate P. aeruginosa from our ICHE study. What if we waited to start interventions to control our peak in summertime pseudomonal infections until September (Intervention B)? I could then ride the epi curve to glory each fall as I reduced infections by 28%. I could then publish my findings and would be asked to write SHEA guidelines recommending what you should do. On the other hand, what if I tried to get ahead of things every spring and start intervening in May (Intervention A)? What if pseudomonal infections went up 5% over the next three months? In that case, I would be told my interventions didn't work, I wouldn't publish my findings and you'd certainly never let me write a SHEA guideline.

With that long background, I'm excited to report that our findings of summer season and higher temperature associated increases in Gram-negative pathogens have been validated in a recent PLoS One paper by Frank Schwab and colleagues.  In a cohort of patients from 73 ICUs in 41 German hospitals covering years 2001-2012, they examined the monthly incidence of 103,000 Gram-positive isolates (S. aureus, Coagulase negative staphylococci (CoNS), E. faecalis and faecium, S. pneumoniae) and 87,000 Gram-negative isolates (E. coli, P. aeruginosa, K. pneumoniae, E. cloacae, S. maltophilia, S. marcescens, Citrobacter spp., A. baumannii) and their relationship to the ambient temperature in the month isolated and also in the prior month.

They found that 11 of the 13 pathogens had a significant temperature association. Only E. faecalis and S. marcescens were not effected by temperature. All remaining Gram-negative pathogens (and CoNS) were positively associated with temperature, and the strongest correlation was with temperature in the prior month. Thus, higher temperatures = higher incidence of Gram-negative pathogens. The magnitude of the effect was similar to what we reported earlier. For example, we reported a 46% increase of E. cloacae in summer vs. winter while they reported a 43% increase. They also found that S. aureus, E. faecium and S. pneumoniae were more frequent when temperatures were colder.

So, as you are reading an outbreak investigation or listening to SHEA2014 talks this week in Denver, ask yourself "did the authors consider seasonal or temperature variation in their analysis?" And if the answer is no, tread carefully. The authors may have ridden the epidemic curve to glory, but you might not be so lucky if you follow their recommendations.

CDC versus Google Trends

Influenza is capturing a lot of media attention this week. Boston declared a public health emergency, activity is widespread in all but two states, and I’ve heard reports of sporadic shortages (of masks, lab testing supplies, pediatric oseltamivir preparations, etc.). Meanwhile, Google Flu Trends (which tracks flu activity via internet searches) has been “blinking red” for several weeks. This Slate piece asks an interesting question: should we be paying more attention to the Google data, or….should we have paid more attention to it as the trend line started ramping up in late November? I suppose the answer depends upon what we could have done earlier to blunt the impact of what the CDC predicts to be a “moderately severe” flu season. Push the vaccine more aggressively, and otherwise get a head start on local preparation, including supplies of masks, antivirals, vaccine, etc.? Promote social distancing or provide advance warning against presenteeism (both at work and school)?

I can’t conclude this post without pointing out that our colleague, Phil Polgreen, published his work on influenza prediction using Yahoo search data before the oft-cited Google paper was published. Phil followed this with some crazy-good Twitter work, too. Credit where credit is due, and all that.

Epidemiological surveillance testing is a waste

Only last month, we posted on the Washington state pertussis outbreak. Back then, there were 640 reported cases through March.  A month later there are 1284 total cases, up from 128 the prior year.  To me, these seem like important data. For one, we've used them to sound the alarm over low vaccination rates.  Now an article in today's NY Times highlights the impact that state budget woes have had on the public health infrastructure and how this has blunted the epidemic response.

Skagit County's (pop: 117,000) top medical officer, ER doc Dr. Howard Leibrand, has some choice words for pertussis testing, which I've pasted below:

If the signs are there, he said — especially a persistent, deep cough and indication of contact with a confirmed victim — doctors should simply treat patients with antibiotics. The pertussis test can cost up to $400 and delay treatment by days. About 14.6 percent of Skagit County residents have no health insurance, according to a state study conducted last year, up from 11.6 percent in 2008. 

“There has been half a million dollars spent on testing in this county,” Dr. Leibrand said late last week. “Do you know how much vaccination you can buy for half a million dollars?” And testing, he added, benefits only the epidemiologists, not the patients. “It’s an outrageous way to spend your health care dollar.” 

Since antibiotic overuse has no cost or downsides from a public health perspective and we don't need to understand the scope of the epidemic, this is probably cool.

Image source: www.healthheritageresearch.com

What will they think of next?

There's a new iPhone app, Outbreaks Near Me, that allows the user to view maps of occurrences of infectious diseases based on media reports. Users can search by disease or geographic location. This must be a nightmare for germophobes.

Zombie control

In this blog, we focus on controversial prevention issues. About some issues, however, there is no debate. For example, how to control zombie outbreaks. Quick, decisive action is essential. As these authors rightly conclude from their novel mathematical model:

"An outbreak of zombies infecting humans is likely to be disastrous, unless extremely aggressive tactics are employed against the undead."

Now if we only had a rapid screening test…..

Where's George?

There's a nice article in today's NY Times about Dirk Brockmann's group at Northwestern, and how they have used data from the "Where's George" currency tracking project to inform their models for spread of infectious diseases, including the 2009 H1N1.  Dr. Brockmann spoke about his approach at SHEA 2009 in San Diego, and it was a terrific and informative talk.